March 12th: A User-Centered Approach

Announcements

• ENGL Majors Days--Undergraduate Day
  Wednesday, 3/20/2013 from 10:00 - 5:00 pm
  Various Places in Fretwell
• ENGL Majors Days--Graduate Day
  Wednesday, 3/22/2013 from 2:30 - 4:30 pm
  Various Places in Fretwell

Plan of Attack

• Midterm Review
• Johnson's Chapter 6
• Bibliographic Essay--5181 students only
• Persona Research--Think about the types of audiences you may serve and technological literacy
• User Document #1 issues

Midterm Review

The midterm is closed in Moodle, but I have a copy up on the big screen. Let's go over the questions.

Questions for Assessing User Complexity Threshold

Here are some questions to ask to figure out what level of complexity your users will tolerate:

• What do users need to know to accomplish goals?
• What are the "basic assumptions" for ideal users?
• To what culture/s will the documents be oriented?
• What level of diction can I expect users to be familiar?
• What are my primary and secondary audiences?
• What are the user contexts?
• What is the users' instant gratification threshold?
• What are the stakes for the users?

Non-biased Questions and Likert Scale

I have a bias. Yes, I, too, am not above bias. I know what you're thinking, "Toscano is biased?!? Say it ain't so!" Well, I am biased against surveys. I think that most questionnaires are highly inaccurate. Self-reporting doesn't yield the most "truthful" results. I also am biased about statistics--I think they're bogus 4 out of 5 times, but that's another story. There are legitimate uses for statistics, but they must be scrutinized and not assumed to be "facts speaking for themselves"--someone always interprets the data to make it useful information. That doesn't mean someone is manipulating an audience, but we ought to think critically about statistics thrown at us.

As individuals interested in getting information from users, we ought to be sensitive to getting that information effectively. I think our discussion might help.

Are yes/no questions effective for getting feedback?

What's the advantage (and disadvantage) of open-ended questions?

Remember, users like to please the testers, so you ought to avoid some leading questions. (I know, a generalization, but 59.8% of what I've read on the subject mentions this...)

What's the difference between the following questions?

• Do you like the interface?
  How do you feel about the interface?
• Isn't this a cute little button to use for submitting?
  How did you use the submit feature? Why? Why not?
• How well did you like the documentation?
  Was the documentation useful? {easy...it's yes/no}
  How did you use the documentation for assistance? Why did you use the documentation for assistance? Why did you not use the documentation for assistance?
  -Even better-
  What kinds of assistance/help would you prefer to be in the documentation?

Paper-Prototyping Groupwork

This only took, well, forever to get to, but I think we can do the "Paper" Prototyping groupwork I wanted you to do weeks ago.

User Doc #2 User Testing

Next week you'll have a more in-depth user test to accompany your more in-depth User Doc #2. I know all of you will work hard to help your fellow classmates by being as objective as possible. The first change, though, will be that I want you all to come up with three personas for the instrument you'll document. I also want you to include the following in what you'll turn in to me on 3/26:

• Describe the instrument

• Explain how the user will approach the set of instructions {online, printed out, on the moon, etc.}

• Plan how you will test your draft

  • Develop a pre-test briefing strategy: make a statement to tell each user (a script), ask questions about comfort level, and/or have them watch something (this last one works best for descriptive documents like natural or mechanical proceses).
  • Come up with five post-test questions that use a Likert scale and have a comments section
  • Set at least four goals and make sure they're measurable
    • User did all steps in XX minutes...
    • User completed the instructions in XX minutes...
    • User used the help menu less than XX times
      -or-
      User never had to look at the help menu or ask for help.
• Describe three personas you had in mind when creating the document--these are Cooper personas and not the detailed ones for your Persona Research assignment
• Come up with five post-test questions that use a Likert scale and have a comments section (just ask the user to comment)
• Set at least four goals and make sure they're measurable

Note: measurable goals mean that users can not only accomplish tasks but they can accomplish them within or under quantifiable standards. For instance, a vague, immeasurable goal would be, "create a user friendly set of instructions"; whereas, a measurable (operationalized) goal would be, "create a document that gets users to setup the instrument within five minutes with no more than two lookups in the help menu." Additionally, a Likert scale could be an attempt to quantify qualitative data: On a scale of 1 to 5 (5 being the highest), please rate your satisfaction with the product."

But let's be critical of our Likert Scales and quantifying subject data in general.

User Doc #2 Work

Next week, I'll be coming around to note who has their drafts. Use this time wisely. Make sure you have post-test questions. Don't have those...get moving on that!

Before We Go...

Next week we'll start Degani's Taming HAL, an interesting work. Your User Document #2 user test is next week (3/19), and it's due in two weeks (3/26).

..